NSERC CREATE Summer School in Atmospheric Science July 23-27, Nottawasaga Inn, Ontario Ice-core studies of atmospheric change in the Arctic

A few examples, and some challenges associated with these studies

Christian M. Zdanowicz Geological Survey of Canada & University of Ottawa Information obtained from ice cores has been a keystone of research on

climate change and the CO2-greenhouse warming over recent decades.

Graphics: UNEP What we learn from studying polar ice cores : 1- Climate history: -Air temperature -Snowfall amounts -Windiness and wind patterns -Relative humidity

2- Air composition history: (Pollutants / climate forcings)

-Trace gases (example: CO2) -Aerosols: dust, sea spray, volcanic ash, sulphate, etc. -Organic compounds: Fatty acids, PAHs, POPs, etc. -Radionuclides: 14C, 10Be, 137Cs, etc. -Biological indicators: pollen, bacteria, DNA fragments, etc. Photo: R. Koerner Ice cores are preferably obtained from near the ice divide of ice sheets, where ice velocity is essentially vertical (no horizontal component) and the ice is thickest

Source: UNEP/GRID Arendal Maps and graphics library The beginnings: Early ice-coring efforts in northwest Greenland, 1956-57.

Langway 2008 CRREL report TR-08-1 A recent deep ice-coring project: EPICA, Dome C, East Antarctica Deep ice-coring sites in Greenland (first and most recent)

NEEM (14 countries Camp Century (USA) incl. Canada) Completed in 1966 Completed in 2010 Depth reached: 1391 m Depth reached: 2538 m Time: ~50,000 years Time: > 150,000 years NEEM

North GRIP (Denmark, France, Switzerland, others) Completed in 2004 Depth reached: 3085 m Time: ~125,000 years

GISP - Dye 3 (USA) Completed in 1981 GRIP (Denmark + Depth reached: 2037 m France + Switzerland) Time: ~30,000 years and GISP2 (USA) Completed in 1992-93 Depth reached: 3029 m Time: ~100,000 years Deep ice-coring sites in Antarctica (first and most recent) Eastern Dronning Maud Land (EPICA) Completed in 2006 Depth reached: 2774 m Time: ~300,000 years Dome Fujii (Japan) Completed in 2006 Depth reached: 3029 m Time: ~720 000 years

Vostok (France/Russia) Byrd Completed in 1996 Depth reached: 3623 m Time: ~400,000 years

Dome Concordia (EPICA) Completed in 2004 Depth reached: 3270 m Time: ~800,000 years

Byrd Station (USA) Completed in 1968 Depth reached: 2164 m Time: ~80,000 years Ice-coring sites in Arctic Canada

(1972-2005) A Agassiz W Prince of Wales

D Devon

Barnes Penny

The larger Canadian Arctic ice caps are partial remnants of the last glaciation The limited thickness of Canadian Arctic ice caps (~100-500 m) imposes limits on the temporal resolution that can be attained in ice-core records Another issue is that surface melt in summer, and water percolation, on many small ice caps can compromise the integrity of climate and atmospheric signals recorded in the snow and firn. The challenge of developing appropriate "transfer functions" to relate proxy climate / atmospheric indicators to actual atmospheric variables. Adequate transfer functions for many aerosols are yet to be developed. The ice-core "paleothermometer": Stable isotopes of O and H in water ice The pioneering Danish physicist Willi Dansgaard, who "invented" ice-core paleoclimatology.

Dansgaard 2005 Frozen Annals

Holocene interglacial

Last glacial period 10 C

The succession of Late Pleistocene glaciations as recorded by water isotope ratios in the EPICA core (E. Antarctica) Arctic Ocean

Spatial variations of 18O in snow across the Queen Elizabeth Islands, Canadian High Arctic. The 18O varies from isotopically "warmer" (heavier) values near Baffin Bay to "colder" (lighter) values in the interior highlands of Axel Heiberg and Ellesmere Islands. Oxygen isotopes can be used as an air mass tracer. Koerner 1979 J. Glaciol. 22. Atmospheric gas “entrapment” in polar firn and ice: A gradual process that can take centuries or millennia in Greenland and Antarctica. Occluded air in deep ice core samples: A record of past atmospheric trace gas mixing ratios

Source: Oregon State University (E. Brook) Occluded air is extracted by dry sublimation and cold-trapping, and fed into a gas chromatograph

Source: Oregon state university (E. Brook) To investigate air (gas) composition changes in the more recent past (last few centuries to millennia), air trapped in firn layers can be pumped out from boreholes and bottled for analysis.

Picture taken at Dome C, East Antarctica http://badc.nerc.ac.uk/data/firetracc/firetracc_help.html#firn The radioactive fallout from the 1962-63 surface thermonuclear bomb tests (pre-moratorium) provides convenient tracers that can be used to determine the rate of burial of firn layers, as well as the age of the gas trapped in it. Reconstructing past atmospheric trace gas mixing ratios from measurements of trapped air in polar firn. This example is from Devon ice cap in the High Arctic. Blue = age structure of firn based on decay of 3H. Red = age structure of trapped 14 air based on decay of C in C02. The two curves are mismatched because CO2 diffuses in firn over time. Clark et al. (2007) J. Geophys. Res. v. 112 (D01301) 3 14 By tracking the rate of CO2 diffusion in firn using radioactive tracers ( H and C), it is possible to model and back-diffuse depth profiles of trace gas mixing ratios measured in trapped air from boreholes, and from there, to reconstruct historical changes in trace gases of relevance to radiative forcing.

Clark et al. (2007) J. Geophys. Res. v. 112 (D01301)

For more information, see also: Firn Record of Trace Gases Relevant to Atmospheric Chemical Change over 100 yrs (FIRETRACC) http://badc.nerc.ac.uk/data/firetracc and special issue of Atmos. Chem. Phys. (2011; papers now under review in ACPD) A (probably far too simple) view of the processes by which aerosols are deposited and preserved in polar snow an ice.

Processes at the air-snow interface

Domine, F. and Shepson,P. (2002) Science 297:1506-1510 Snow ----- Firn ----- Ice metamorphism (Water vapor and heat transfer) Domine,F. and Shepson,P. (2002) Science 297:1506-1510 Processes that can modify the chemical composition of ice (at the ice grain scale) Domine,F. and Shepson,P. (2002) Science 297:1506-1510 The main air mass transport pathways for pollution entering the Arctic.

AMAP Arctic Air Pollution 2006 An example of the geographic "footprint" of pollution in the Arctic: Estimates of the dispersion and deposition of lead (Pb) from industrial sources. Note the strong Pb concentrations originating from the Kola Peninsula in NE Russia. Actual Pb concentrations in Arctic snow measured across the pole, showing the strong gradient from the Russian to Canadian side, and which matches the pattern of Pb dispersion in air from the previous slide. Some of the locations from which ice-core archives of Arctic air pollution have been developed.

Agassiz ice cap Prince of Wales Devon icefield ice cap

Penny central ice cap Greenland

AMAP Arctic Air Pollution 2006 An early study revealed rising acidity levels in Arctic snow in the second half of the 20th century, resulting from acidifying gases such as SOx an NOx. declining SO4 levels

Trends in monthly-averaged sulphate and nitrate concentrations measured in March and April at eight Arctic air monitoring stations over the period 1982-2003: A steady decline is seen, resulting from better pollution abatment measures after the 1970s. After Quinn et al., Tellus vol. 59B (2007) Red = North American SO2 emissions, Blue = Eurasian SO2 emissions.

Goto-Azuma and Koerner (2001) J. Geophys. Res. vol. 106 Geographic variations in changing sulfate deposition rates across Greenland 8 7 6 5 Acidity 4 (M) 3 2 1 0 AD 1550------> 2000

Sulfate deposition peaked in the mid-1970s and declined afterwards, but the timing and rates of increases / decline vary geographically across Greenland, reflecting the influence of different air mass transport pathways and "airsheds".

Pasteris, McConnell et al. 2010 Declining trends in sulphate deposition in the southern Baffin Island region, as recorded in cores from Penny ice cap, Cumberland Peninsula. Rates of atmospheric Pb deposition in Greenland (bottom) and on Devon ice cap in the Canadian High Arctic (top) over the past few hundred years, from ice-core measurements. Looking at a very different place: Mount Logan in the SE Asia southwestern Yukon... industrial centers Mount Logan

northern China + Mongolia deserts

AMAP Arctic Air Pollution 2006 The ice-core record of atmospheric Pb deposition in the SW Yukon is strikingly different from that in Greenland, and reflects increasing trans- Pacific transport of polluted air from East Asia. The Mt Logan ice core comes from a site at 5400 m, in the mid-troposhere, and "captures" aerosols being transported in fast winds at that altitude from Asia.

Osterberg et al. (2008) GRL v. 35 Another type of "natural" pollutant that regularly contaminates the northern hemisphere: Desert dust from the arid regions of northern China and Mongolia that blows across the Pacific every spring. Here, measurements of aluminum (a tracer for dust) in aerosols in the mid-Pacific atmosphere. An example of a long-range transport event: The April 2001 Asian dust outbreak

Peak storm period, Baichen , N. China, April 7, 2001 Zhenbeitai, China, 9 April 2001 Photo by Zev Levin, Courtesy Xinhua Press Service Photo: Sondra Sage, ACE-Asia Program April 9: Dust cloud emerging April 14: Dust cloud approaches from the Gobi desert W coast of N. America Images from U.S. Naval Aerosol, Analysis and Prediction System courtesy D. Westphal, U.S. Naval Research Laboratory

Hazy sunset over Toronto skyline, April 17, 2001 Asian Desert Dust on Mount Logan, yukon: An estimated 6000 tons of Asian desert dust fell in this region over the course of a few days/

Dusty snow layer (thickness ~8 cm) dust dust in recognizable snowis as far Greenland. as "fingerprint"desert dust and locate itssource region. The sign Isotopictravers such as the ratios Srand of Nd isotopes can be atureAsian of used to

Zdanowicz et al. (2006) Geochim. Cosmochim. Acta vol. 70 Very large particles of desert dust found in snow at 5400 m on Mt Logan testify to the extreme transport efficiency of mid-tropospheric winds blowing across the North Pacific.

Zdanowicz et al. (2006) Geochim. Cosmochim. Acta vol. 70 An example of using ice-core measurements used to estimate the radiative forcing of aerosols (here: soot) in the Arctic.

Left: Changes in black carbon (BC; black) and excess sulphur (red) deposited in west-central Greenland since AD 1773. Right: Modelled surface radiative forcing (RF) resulting from the BC loadings measured in ice cores. The RF shown is for early summer (June-July), using total BC (black), and estimated industrial BC (red) only.

McConnell et al. (2007) Science 317: 1382-1384. Existing ice-core records of atmospheric Hg The growing interest in the chemistry of halogens, ozone and mercury (Hg) in the Arctic has led to new efforts to obtain ice-core records of Hg deposition from sites in the mid-latitudes and the polar latitudes. Existing ice-core records of atmospheric Hg variations in the Northern Hemisphere: Not two are the same !! Lots of work for atmospheric scientists to explain these differences....

CARA national report on Hg pollution, in preparation Some of the key papers cited in the lecture:

Barrie, L. et al. (1985) 20th century trends in Arctic air pollution revealed by conductivity and acidity observations in snow and ice in the Canadian High Arctic. Atmos. Env. v. 19: 2055-2063.

Dominé, F. and Shepson, P. (2002) Air-Snow interactions and atmospheric chemistry. Science v. 297: 1506-1510.

Clark, I. et al. (2007) CO2 isotopes as tracers of firn air diffusion and age in an Arctic ice cap with summer melting, Devon Island, Canada. J. Geophys. Res. v. 112, D01301.

Goto-Azuma, K. and Koerner, R. (2001) Ice-core studies of anthropogenic sulfate and nitrate trends in the Arctic. J. Geophys. Res. v. 106, D5, 4959-4969.

McConnell, J. et al. (2007) 20th-century industrial black carbon emissions altered Arctic climate forcing. Science v. 317: 1382-1384.

Osterberg, E. et al. (2008) Ice core record of rising lead pollution in the North Pacific atmosphere. Geophys. Res. Lett. v. 35, L05810.

Pasteris, D. et al. (2010) Anthropogenic acid deposition in a southern Greenland core. Presented at the State of the Arctic conference, Miami.

Quinn, P.K. et al. (2007) Arctic haze: current trends and knowledge gaps. Tellus 59B: 99-114.

Zdanowicz, C. et al. (2006) Asian dustfall in the St-Elias Mountains, Yukon Territory. Geochim. Cosmochim. Acta 70: 3493–3507. Suggested further readings:

Langway, C.C., Jr. (2008) The history of early polar ice cores. Cold Regions Research and Engineering Laboratory Technical Report TR-08-1.

Dansgaard, W. (2005) Frozen Annals: Greenland Ice Cap Research. Niels Bohr Institute, Copenhagen, Denmark, 124 p. [The late Willi Dansgaard's fascinating memoir on his carreer as a pioneer of ice-core studies.]

Elsevier's Encyclopedia of Quaternary Science (2007): Ice core studies, pp.1145- 1311. [This resource contains short chapters reviewing various aspects of ice-core research techniques].

Web resources: http://www.climate-expeditions.org [A new educational web site hosted by the US ice drilling program: it provides a launching point with links to many of the World's leading ice core research laboratories and programs.]